=Paper=
{{Paper
|id=Vol-1082/preface
|storemode=property
|title=None
|pdfUrl=https://ceur-ws.org/Vol-1082/01-preface.pdf
|volume=Vol-1082
}}
==None==
<pdf width="1500px">https://ceur-ws.org/Vol-1082/01-preface.pdf</pdf>
<pre>
                                  Preface
Data Mining on Linked Data Workshop (DMoLD'13) with Linked Data Mining
Challenge was held on 23 September 2013 in Prague, Czech Republic. It was
part of ECML/PKDD2013, the European Conference on Machine Learning and
Principles and Practice of Knowledge Discovery in Databases. It was the rst
workshop in a series foreseen for researchers and practitioners interested in `KDD
rst'.
   Linked data (LD), published on the web in RDF format, has been so far
nearly untouched by advanced data mining methods. Due to the nature of data
it poses specic challenges. There is a larger variety of link types in the graph.
Presence of semantic links between datasets makes data more complex. It also
allows inclusion of new linked datasets into the mining dataset nearly on the
y, making the feature selection problem extremely hard. Moreover, there are
heterogeneous data models, with varying degrees of completeness and credibility.
The potential of interlinking to external datasets is tempting. The series should
provide forum for discussions on adapting techniques from KDD rather than
inventing LD-tailored approaches from scratch.
   The workshop consisted of a keynote speech, an open track, a challenge track,
and a discussion panel. The Program Committee consisting of 13 members ac-
cepted two papers for the open track and two for the challenge track. As the
workshop was foreseen mostly for KDD community it featured the challenge
event addressing a real business KDD problem rather than problems specic
to Semantic Web data handling. In 2013 the core of the Linked Data Mining
Challenge (LDMC) held as part of the workshop was a dataset from the highly
topical public procurement domain.
   Heiko Paulheim gave a keynote talk on Exploiting Linked Open Data as
Background Knowledge in Data Mining.
   The workshop was supported by the grant from European Union's 7th Frame-
work Programme provided for the project LOD2 Creating Knowledge out of
Interlinked Data (GA no. 288176).




August 2012                                                     Claudia d'Amato
                                                                      Petr Berka
                                                                  Vojt¥ch Svátek
                                                                 Krzysztof W¦cel
                    Conference Organization
Programme Chairs
Claudia d'Amato, University of Bari, Italy
Petr Berka, University of Economics, Prague, Czech Republic
Vojt¥ch Svátek, University of Economics, Prague, Czech Republic
Krzysztof W¦cel, Pozna« University of Economics, Poland



Program Committee
Jose María Alvarez-Rodríguez    South East European Research Center, Greece
Sören Auer                      Universitat Leipzig, Germany
Bettina Berendt                 KU Leuven, Belgium
Marko Grobelnik                 JSI, Ljubljana, Slovenia
Fazel Famili                    NRC Institute for Information Technology Ot-
                                tawa, Canada
Nicola Fanizzi                  University of Bari, Italy
Agnieszka Šawrynowicz           Poznan University of Technology, Poland
Pablo Mendes                    Free University Berlin, Germany
Martin Ne£aský                  Charles University Prague, Czech Republic
Zbigniew Ra±                    University of North Carolina, Charlotte, USA
Jerzy Stefanowski               Poznan University of Technology, Poland
Volker Tresp                    University of Munich, Germany
Djamel Zighed                   University Lumiere, Lyon, France



Additional Reviewers
Domain experts in public contract data analysis:


  Jana Chvalkovská, Charles University, Prague
  Ji°í Skuhrovec, Charles University, Prague
                                  Table of Contents
Invited Talk
Exploiting Linked Open Data as Background Knowledge in Data Mining .
    Heiko Paulheim



Open Track
Lattice Based Data Access (LBDA): An Approach for Relating Data
and Linked Open Data in Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
    Mehwish Alam, Melisachew Wudage Chekol, Adrien Coulet, Amedeo
    Napoli and Malika Smail-Tabbone


A Fast and Simple Graph Kernel for RDF                     ...........................
    Gerben Klaas Dirk de Vries and Steven de Rooij



Challenge Track
Linked Data Mining Challenge (LDMC) 2013 Summary . . . . . . . . . . . . . . . .
    Vojtech Svatek, Jindrich Mynarz, Petr Berka


Graph Kernels for Task 1 and 2 of the Linked Data Data Mining
Challenge 2013        ...................................................
    Gerben Klaas Dirk de Vries


A Machine Learning approach to the Linked Data Mining Challenge 2013
    Eneldo Loza Mencia, Simon Holthausen, Axel Schulz and Frederik Janssen

</pre>